Estimation of Ligand Binding Free Energy Using Multi-eGO

Area of Science:

• Computational chemistry and molecular modeling
• Biophysics and structural biology

Background:

• Accurate prediction of ligand-protein binding is crucial for in silico drug design.
• All-atom molecular dynamics (MD) simulations provide mechanistic insights but are computationally expensive.
• Coarse-grained models offer speed but often sacrifice accuracy and resolution.

Purpose of the Study:

• To introduce and evaluate the multi-eGO atomic model for efficient and accurate binding free energy estimation.
• To address the speed-accuracy trade-off in molecular dynamics simulations for ligand binding.
• To demonstrate the utility of multi-eGO in various ligand-protein binding scenarios.

Main Methods:

• Application of the multi-eGO atomic model for binding free energy calculations.
• Utilized thermodynamic integration and metadynamics simulations.
• Performed single molecule simulations and explicit ligand titration for concentration-dependent binding.

Main Results:

• Accurate binding free energy calculations for benzene-lysozyme, dasatinib/PP1-Src kinase, and 10074-G5-Aβ42.
• Revealed multiple binding/unbinding pathways for benzene using multi-eGO.
• Demonstrated comparable accuracy to traditional methods with significantly reduced computational cost.

Conclusions:

• The multi-eGO atomic model significantly reduces the computational cost of accurate binding free energy calculations.
• This model shows great potential for developing and benchmarking in silico ligand binding techniques.
• Enables more efficient and reliable virtual screening and drug design processes.